1 The EU-project SMAS: Safety Monitoring and Assurance

1

1

The EU-project SMAS: Safety Monitoring and Assurance System for

Chilled Meat ProductsPetros S. Taoukis

International Workshop

Quality Management of the Chill Chain

Athens, GREECE, 16 December, 2005

National Technical University of Athens, School of Chemical EngineeringLaboratory of Food Chemistry and Technology

2

2

S M A S

Development and application of a TTI based Safety Monitoring and Assurance System for Chilled Meat Products

QLK1-CT-2002-02545

A European Commission Research and Technology Development Project

FIFTH FRAMEWORK PROGRAMMEQuality of life and management of living resources

http://smas.chemeng.ntua.gr

3

3Meat Chill Chain- Need for better management Meat products are perishable and unless processed, packaged,distributed and stored appropriately can spoil in relatively shorttime. Overgrowth of incidental pathogenic bacteria like Listeria monocytogenes, Salmonella sp. and Escherichia coli followed by undercooking or inadequate preparation may pause a potentialhazard for the consumer. Despite the proliferation of food safetyregulations and the application of safety management systems suchas HACCP, risk assessment studies show that foodborne disease hasremained a main concern in the last decade.

Why SMAS?

4

4

Meat Chill Chain- Need for better management

It is generally recognized by the European industry, retailers,food authorities and even consumers that the weakest link thataffects directly safety and quality of chilled products is theactual chill chain. A big percentage of foodborne disease isdue to temperature abuse.

Why SMAS?

5

5Meat Chill Chain- Need for better management Application of an optimised quality and safety assurancesystem for chilled distribution of fresh meat and meat productsrequires continuous monitoring and control of storageconditions from production to consumption. The systematicmanagement of the chill chain and the improved evaluation ofsafety, quality and shelf life of meat can lead to reduced safetyrisk and increased quality, with a significant health and economic impact to the European society and market.

Why SMAS?

6

6

Wholesaler

RetailerHospital

Schools... Producer

Packer/Processor

Chill chain: Monitoring andChill chain: Monitoring and ManagementManagement

Consumer

From packing to consumer

7

7What is SMAS?

SMAS is an integrated chill chain management system,expected to lead to an optimised handling of products in termsof both safety and quality. It is based on the ability tocontinuously monitor the storage conditions of each productwith the use of Time Temperature Integrators (TTI). TTI are inexpensive “smart labels” that show an easilymeasurable, time and temperature dependent change thatcumulatively reflects the time-temperature history of the foodproduct. TTI response can be correlated to meat safety andquality status at any point of the distribution chain providingan effective decision tool. Project QLK1-2002-02545

8

8

The SMAS project The acronym SMAS summarizes the long title of the 3 year (2003-2006) action project “Development and application of a TTI basedSafety Monitoring and Assurance System for Chilled MeatProducts”, co-ordinated by the National Technical University of Athens (NTUA). Funded by the EC, it is part of the key action ofFood, Nutrition and Health. The project basis consists of validatedpredictive models of predominant meat pathogens growth andkinetics of the response of selected TTI, all applied in an expanded TTI application scheme that translates TTI response to meatmicrobiological and quality status. 7 Institutes/Companies are members of the SMAS project,working on its 6 main interrelating workpackages with the ultimate purpose to deliver an effective chill chain decision andmanagement tool.

9

9

OVERALL OBJECTIVE

State of the art ofTTI technology + Quantitative risk assessment

⇓Development of SMAS, an effective and reliable safety assurance and

quality optimization management system for meat products, extending from production to the table of consumer

SMAS objectives

10

10

SMAS validation& application

In meat industry

Consumer perception & benefit

Microbial modelling

RiskAssessment

(dose-response, data on prevalence/concentration)

TTI Development, modelling and

optimization

SMAS Development,

SMAS

application

What is the structure of SMAS?

11

11PROJECT WORKPLANWorkpackages & their interrelation

Project QLK1-2002-02545

Development and Application of a TTI BasedSafety Monitoring and Assurance System (SMAS)

for Chilled Meat Products

WP1MicrobialModeling

WP2Risk

Assessment

WP3TTI

DevelopmentModeling &

Optimisation

WP4SMAS

Development

PRODUCTS TTI SMAS

WP5SMAS Validation

and Application inMeat Industry

WP6ConsumerPerception& Benefit

APPLICATION

12

12 The major expected achievements of the project will be: • Accurate, validated mathematical models for safety and quality related

microorganisms of ready to cook meat products. They will provide themeat industry with a tool for product development and safety assuranceand the European authorities with a quantitative means for meatproduct risk evaluation.

• The development and study of an assortment of Time TemperatureIntegrators (TTI) suitable for meat safety monitoring. These TTI willprovide the meat industry and retail business with effective tools tomonitor the chill chain.

• Improved distribution logistics and management of the meat chillchain from the application the Safety Monitoring and AssuranceSystem (SMAS). SMAS could replace the current “First In First Out”(FIFO) practice and lead to risk minimization and quality optimization.

• Increased ability of the meat sector to control its weak link, the chillchain

.

13

13

Current practice: First In- First Out (FIFO)

Disadvantages: ignores variations of product characteristics

ignores the REAL time-temperature history of the product

Proposed practice: SMASMain Advantages:

variations of product characteristics are consideredthe REAL time-temperature history of the product is

taken into account based on TTI response

SMAS vs FIFO

14

14

TThhee ccoonnttrriibbuuttiioonn ooff SSMMAASS iinn tthhee cchhiillll cchhaaiinn mmaannaaggeemmeenntt ccaann bbee vviissuuaalliizzeedd aass aa mmiinniimmiizzaattiioonn ooff rriisskk ffoorr iillllnneessss aanndd ooppttiimmiissaattiioonn ooff tthhee mmeeaatt pprroodduucctt qquuaalliittyy aatt tthhee ttiimmee ooff ccoonnssuummppttiioonn

15

15

05

10152025303540

FIFOFIFO

One

out

of

a tri

llion

One

out

of

10 b

illio

ns

One

out

of

100m

illio

ns

One

out

of

a m

illio

n

One

out

of

10.0

00

One

out

of

100

% o

f pro

duct

s

Probability of illness

16

16

05

10152025303540

FIFOFIFO

One

out

of

a tri

llion

One

out

of

10 b

illio

ns

One

out

of

100m

illio

ns

One

out

of

a m

illio

n

One

out

of

10.0

00

One

out

of

100

% o

f pro

duct

s

0

10

20

30

40

One

out

of

a tri

llion

One

out

of

10 b

illio

ns

One

out

of

100m

illio

ns

One

out

of

a m

illio

n

One

out

of

10.0

00

One

out

of

100

SMASSMAS


17

17

One

out

of

a tri

llion

One

out

of

10 b

illio

ns

One

out

of

100m

illio

ns

One

out

of

a m

illio

n

One

out

of

10.0

00

One

out

of

100

% o

f pro

duct

s


05

10152025303540

FIFOFIFO

SMASSMAS%

of p

rodu

cts


18

18

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5

FIFOFIFO

REJ

ECTE

D

Time to end of shelf life (days)

% o

f pro

duct

s

Product quality at consumption

19

19

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5

FIFOFIFO

REJ

ECTE

D


% o

f pro

duct

s

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5

FIFOFIFO

REJ

ECTE

D


% o

f pro

duct

s

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5


% o

f pro

duct

s SMASSMAS

REJ

ECTE

DProduct quality at consumption

20

20

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5.0Time to end of shelf life (days)

% o

f pro

duct

s SMASSMASFIFOFIFOR

EJEC

TED


21

21

SMAS BROCHURE

22

22

TasksTasksStudy of Study of the temperature conditions in the chill chain the temperature conditions in the chill chain (fluctuations during transport/storage, variability within (fluctuations during transport/storage, variability within the domestic equipment)the domestic equipment)

CorrelationCorrelation temperature handling to food quality with the use temperature handling to food quality with the use

of of Time Temperature IndicatorsTime Temperature Indicators

Validation of the Validation of the Safety Monitoring and Assurance System (SMAS) Safety Monitoring and Assurance System (SMAS)

by simulation and experiment by simulation and experiment

23

23

CHARACTERISTICS OF THE ACTUAL CHILL CHAIN

24

24

Electronic dataloggersWeak links in the chill chain

1st stage of the survey:

Data loggers were sealed in packages of ground beefand monitored from processing through distribution to delivery and storage in SuperMarkets all over Greece

Survey for temperature conditions (1)

25

25

02468

101214

0 5000 10000 15000

time (min)

tem

pera

ture

0

2

4

6

8

10

12

0 5000 10000 15000 20000

time (min)

tem

pera

ture

Temperature conditions FROM production TO the retail outlet

26

26

0

2

4

6

8

10

12

14

0 2000 4000 6000 8000 10000

tem

pera

ture

time (min)

time (min)

02

46

810

1214

0 5000 10000 15000 20000


27

27

-202468

101214

0 5000 10000 15000 20000

tem

pera

ture

time (min)

0

5

10

15

0 5000 10000 15000 20000te

mpe

ratu

retime (min)


28

28

-10

-5

0

5

10

15

0 5000 10000 15000

tem

pera

ture

time (min)

-5

0

5

10

15

0 5000 10000 15000te

mpe

ratu

retime (min)


29

29te

mpe

ratu

re

-1

0

1

2

3

4

4000 6000 8000 10000 12000

time (min)

-1

0

1

2

3

4

4000 6000 8000 10000 12000

time (min)te

mpe

ratu

re


30

30

Conclusions from survey (1)

• Sharp (but short) increases of temperature (during transport)• Cases of retail storage at temperatures > 7°C• Significant temperature fluctuations


31

31

2st stage of the survey:Variations WITHIN the domestic refrigerator

Survey for temperature conditions (2)

Electronic dataloggers

4 data loggers were distributed randomly to potential consumers to monitor variations

INSIDE the refrigerator

1

2

3

4 1: upper shelf2: middle shelf3: lower shelf4: door

32

32

-5

0

5

10

15

0 1000 2000 3000 4000

UP MIDDLEDOWN DOOR

1.52.53.54.55.56.5

50 550 1050 1550 2050 2550 3050

time (min)

time (min)

tem

pera

ture

tem

pera

ture

Temperature conditions IN the domestic refrigerator

33

33

0

5

10

15

0 1000 2000 3000 4000

UPMIDDLEDOWNDOOR

tem

pera

ture

time (min)

0

5

10

15

1000 1500 2000 2500 3000 3500 4000 4500 5000

time (min)te

mpe

ratu

re

Temperature conditions IN the domestic refrigerator

34

34

02468

10121416

0 100 200 300 400time (min)

tem

per

atu

re

upper shelfmiddle shelflower shelfdoor

02468

10121416

0 100 200 300 400time (min)

tem

per

atu

re

35

35%

of d

omes

tic re

frig

erat

ors

Temperature conditions WITHIN the domestic refrigerator

26%

28%

23%

15%

8%

< 4°C4°C<T< 6°C6°C<T< 8°C8°C<T< 10°C10°C<T< 12°C

Average temperature of domestic refrigerators(~400 cases)

Temperature variation within domestic refrigerators

05

1015202530

-2.0-0.80.31.52.73.85.06.27.38.59.710.812.0

upper shelf middle shelf

lower shelf door

Temperature (°C)

36

36

Temperature variationsin distribution and storage conditions

NEED for continuous monitoringTTI

37

37

TTI PRINCIPLES & APPLICATION

38

38TTI: main principles

Time Temperature Indicators (TTI) are simple, inexpensivedevices that can show an easily measurable, time and temperaturedependent change that cumulatively indicates the time-temperature history of the product from the point of manufacture to theconsumer, allowing the location and the improvement of the criticalpoints of the chill chain

TYPES OF TTI

Polymer basedENZYMATIC

Diffusion based

Microbial

Photochemical

PRINCIPLES OF TTI

39

39

Time Temperature Indicators

The indicator starts with two liquid-filled pouches heat sealed into plastic

After exposure to time and temperature, the contents turn from green to yellow to red

The contents are mixed by bursting the seal between the pouches by pressure

ΤΤΙ

40

40

Tricolour response TTI

Food product shelf life (ts)

Production Expiration

TTI

41

41TTI Configurations

Bicolor TTIsColor change from green to yellow

Tricolor TTIsan initial green color changes into an amber or orange red and ends with a final pure red color, giving a much more clear perception especially to consumers,where the TTI mediates an alarm function in the form of a traffic light

TTI Development & Study

42

42Alternative methods measuring TTI response

Instrumental measurementColormeter such as Minolta CR 200Digital imagers such as a scanner

Visual measurement8 color scale

A simple 3 color scale (consumer use)

Measuring devices for TTI response

43

43TTI RESPONSE KINETICS

Χ: measurable change of ΤΤΙΧ: measurable change of ΤΤΙ

Response functionResponse function:

F(X) = k t

Temperature dependenceTemperature dependence -- expressed by expressed by EaEa

( ) t 11 ⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−

−==

ref

aI TTR

EexpkktXF I

ref

44

44

Kinetic study of enzymatic TTI

45

45

Kinetic study of TTIisothermal experiments

283338434853586368

0 100 200 300 400

0°C 5°C 6.5°C

8°C 10°C 15°C

0.00.2

0.40.60.8

1.01.2

0 50 100 150 200 250 300 350 400

Xc

time (h)time (h)

C

M4-10

46

46

TTI

0.000.200.400.600.801.001.201.401.601.802.00

0.0 100.0 200.0 300.0 400.0 500.0

0°C 5°C6.5°C 8°C10°C 15°C

M4-10

F(X

c)

time (h)

47

47

-6

-5.5

-5

-4.5

-4

-3.5

-3

0.00345 0.0035 0.00355 0.0036 0.00365 0.0037

1/T

lnk

Ea= 76.2 kJ/mol (18.2kcal/mol) R2= 0.972

Arrhenius plot for TTI

M4-10

48

48

0

5

10

15

0 1000 2000 3000 4000 5000

Teff = 8.2°C

F(X

c)

time (h)T

(°C

)

time (h)

M4-10

Validation under nonisothermal conditions

00.20.40.60.8

11.21.41.61.8

0 50 100 150 200

experimentalConf.int.predictionLinear (experimental)

49

49TTI Study & Development

A wide range of TTIs have been developed, kinetically studied & validated under variable temperature conditions

2 TTI configurations:

Alternative methods of measuring TTI response

Different color scales to allow accurate visual TTI readings

Bicolor (green to yellow)

Tricolor (green to yellow to red)

TTI Study

50

50Wide range of TTIs

Different TTI designs of various response characteristics response from hours to several weeks at refrigeration temperatures

The TTIs temperature sensitivity ranged from 50 to 200KJ/mol covering the respective range of bacteria growth in meat products

TTI Study

51

51Time to endpoint of Indicator (h)TTIType

EA

(kcal/mol)–(kJ/mol) 0oC 5oC 10oC 15oC

M4-6 16.9-70.7 210 120 60 35

M4-7 19.0-79.5 250 150 70 50

M4-8 17.0-71.2 280 190 85 50M4-9 21.4-89.6 415 190 74 57

L4-42 35.5-148.6 2500 800 257 85

L4-54 40.0-167.4 3000 980 300 80

B4-2 15.8-66.1 67 42 20 10

C4-13 11.1-46.5 400 260 200 150

L4-58 22.0-92.1 2500 1300 700 220

LM4-8 20-83.7 310 155 80 50

C4-8 10.2-42.7 200 165 122 67

M4-10 18.5-77.4 410 250 120 80

M4-23 16.3-18.2 917 630 300 16

M4-29 24.0-100.5 1000 670 280 170

L4-4 35.8-149.9 264 85 20 8

L4-11 47.9-200.5 880 197 52 11L4-16 36.4-152.4 940 300 70 25

C4-20 12.6-52.7 570 470 300 210

Wide range of TTIs

TTI Study

52

52

0

50

100

150

200

250

0 5 10 15

M4-5L10-1

temperature (°C)

t s(h

)

TTI kinetic results – Comparison Type L vs M

53

53

Temperature

0

100

200

300

400

500

600

0 5 10 15

ListMeat

Pseudomonas

M

L

Time h

54

54

KINETICS OF MICROBIAL GROWTH

Growth for variable temperature distribution:

Τeff : constant temperature that results in the same quality change as the variable temperature distribution over the same time period

( ) ( )[ ]

tRT

Eexp

tdT1

REexp dttTNf

eff

A

t

0

At

0maxt

ref

ref

⎟⎟⎠

⎞⎜⎜⎝

⎛ −µ=

⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎠⎞

⎜⎝⎛−

µ=µ= ∫∫

55

55TTI RESPONSE KINETICS

Χ: measurable change of ΤΤΙΧ: measurable change of ΤΤΙ

( ) t 11 ⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−

−==

ref

aI TTR

EexpkktXF I

ref

Response functionResponse function:

Using effective temperatureUsing effective temperature::

( ) ( )[ ] tdTTR

EexpkdttTkXF

t

ref

aI

t

tI

ref ∫∫ ⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−

−==

00

11

( ) tTTR

EexpkXF

refeff

aIt

I

ref ⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛−

−=

11

For variable temperature distributionFor variable temperature distribution::

56

56

Teff

f (N)Response function Meat microbial

Iref AI E,k ΕA (ΤΤΙ) ? ΕA (food)

Teff

TeffTeff

TTI colormeasurement

TTI colormeasurement

F(X)F(X))

NtNtMeat microbial status at time t

`Response functionGrowth models

Iref AI E,k ΕA (ΤΤΙ) ? ΕA (food)

Requirement

TeffTeff

= Aref E,µ

LagAref E,Lag( ) ( )[ ] td

T1

RE

expk dttTkXFt

0

aI

t

0t

Iref ∫∫ ⎟⎟

⎠

⎞⎜⎜⎝

⎛ −==

( ) tT

1RE

expkXFeff

aIt

Iref ⎟⎟

⎠

⎞⎜⎜⎝

⎛ −=

( ) ( )[ ]

tRT

Eexp

tdT1

REexp dttTNf

eff

A

t

0

At

0maxt

ref

ref

⎟⎟⎠

⎞⎜⎜⎝

⎛ −µ=

⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎠⎞

⎜⎝⎛−

µ=µ= ∫∫

TTI application scheme

57

57

EA (TTI) (kJ/mol)

Teff (°C)(predicted)

∆Teff (°C)Teff (TTI)- Teff

(actual)

Npred/Nactual

46 (Type C) 8.04 -0.570

0.140

2.150

0.062

0.005

76 (Type M)

8.75

0.43

1.03

34.52

0.90

150 (Type L)

10.76

Double TTI

(46-150)(Type C-Type L)

8.67

0.82Double TTI

(46-76)(Type C-Type M)

8.61

Single TTI

∆Teff (°C)

Npr

ed/N

actu

al

0.01

0.1

1

10

100

1000

-4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4

logNlogNoo=2.5, =2.5, logNlogNfinalfinal=8 =8 (E(Ea a ≅≅ 70.3 kJ/mol, Shelf [email protected] kJ/mol, Shelf life@4°°C=305h)C=305h)

CORRECTION of the error in the Teff estimation

58

58INPUT OUTPUT

Food Characteristics

LagArefAref E,Lag,E ,µ

T(t) data

Dataloggerprofile

TTI response• Lab scale

• Visual scale

OR

Teff calculation

Acceptability limit(e.g. Ns for microbial growth),

Initial status (No),Food spoilage model

Remaining Shelf Life

TTI response models

59

59

60

60(2)

(TTI 1)(3)

(TTI 2)

(1)

(RESULTS)

(a) Values for TTI color (Chromameter – Scanner- … Lab scale)

Teff, Shelf life

61

61

SMAS PRINCIPLES & APPLICATION

62

62

Distribution Center

Export marketLocal market

Retail Storage(Super Market)

Transport

Display CabinetDomestic

Fridge

Stock Display

Final product Storage

Consumption

1st Decision pointProduct promotion

2nd Decision point

Initial contamination of L. monocytogenes

0

20

40

60

80

100

<0 0-1 1.0-10 10-100CFU/g

perc

enta

ge

(12h at 4 oC) (8h at 6 oC)

(24h at 4 oC)

Distribution of loca l transporta tion time

0

5

10

15

20

25

2 4 6 8 10 12 14 16 18

Tim e (h)

perc

enta

ge

Distribution of export transportation time

0

5

10

15

20

25

30

20 30 40 50 60 70 80

Time (h)

perc

enta

ge

Temperature distribution f or commercial ref rigerators

0

2

4

6

8

10

12

14

16

18

20

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

te m pe rature (°C)

perc

enta

ge

Distribution of time in domestic refrigerator

0

5

10

15

20

25

30

35

0 15 30 45 60 75 90 105

Time (h)

perc

enta

ge

(4 oC)

(6, 18, 30h)

chill chain scenario

Temperature distribution fordomestic refrigerators

6°C<T< 8°C23%

4°C<T< 6°C29%

8°C<T< 10°C16% < 4°C28%

10°C<T< 12°C4%

Distribution of cooked ham aw

05

101520253035

0.975 0.973 0.971 0.969 0.967 0.965 0.963

aw

perc

enta

ge

63

63

Adistant72 h Domestic StorageRetail Storage

Distribution center1st decision point

??

CookingTransportation

Domestic StorageRetail Storage

÷ 2local B12 h

Cooking

1. Random SplitOR

2. SMAS based split

64

64

Data from:

B

ANt(1)

Nt(2)

Nt(i)

Nt(n-1)

Nt(n)

÷ 2

RANDOM SPLIT

Temperature Distribution Teff (1)

Teff (2)

Teff (i)

Teff (n-1)Teff (n)

Time Distribution

1. Random Split

65

65

Retail Storage

distantDomestic Storage

CookingDistribution center

1st decision point??

Domestic Storage

Cooking

72 h

Retail Storage

local12 h

A

B

0

2

4

6

8

10

0 5 10 15 20 25

Temperature Distribution

Time Distribution

Growth model(Kinetic constants)

Data from:

Teff (1)Teff (2)

Teff (i)

Teff (n-1)Teff (n)

Nt(1)

Nt(2)

Nt(i)

Nt(n-1)

Nt(n)

÷ 2

RANDOM SPLIT

0

5

10

15

20

25

-12-10

.6666

67-9.

3333

333 -8

-6.66

6666

7-5.

3333

333 -4

0

5

10

15

20

25

30

35

-14 -13.3 -12.5 -11.8 -11 -10.3 -9.5 -8.75 -8 -7.25 -6.5 -5.75 -5

Ris

k A

ssess

men

t b

ase

d

on

ra

nd

om

sp

lit

A

B

Transportation

1. Random Split

66

66

Adistant72 h Domestic StorageRetail Storage

Distribution center1st decision point

??

CookingTransportation

Domestic StorageRetail Storage

÷ 2local B12 h

Cooking

1. Random SplitOR

2. SMAS based split

67

67

Nt(1)

Nt(n)

Nt(n-1)

Nt(3)Nt(2)

(1)

(n)

(3)(2)

(n-1)

TTI response Teff 1TTIsoftware

Product’s IDProduct’s kinetic model

A B

Nt’ distributionNmed

B

A

Nt’(i) < Nmed?

TRUE

FALSETTI response Teff n

TTIsoftware

Product’s IDProduct’s kinetic model

SMAS based splitSMAS based split

Nt’(1)Nt’(2)

Nt’(1)<..<Nt’(i)<…<Nt’(n)

÷ 2

Nt’(3)

Nt’(n-1)

Nt’(n)

2. SMAS based split

pH, aw, …

pH, aw, …

68

68distant

Retail StorageDomestic Storage

CookingDistribution center

1st decision point??

Domestic Storage

Cooking

72 h

Retail Storage

local12 h

SMA

S B

ASE

D S

PLI

T

A

B

TTI response Teff (1)TTIsoftware

Product’s ID Product’s kinetic model

TTI response Teff(n)TTIsoftware

Product’s ID Product’s kinetic model

(1)

(n)

(3)(2)

(n-1)

Nt(1)

Nt(n)

Nt(n-1)

Nt(3)Nt(2)

A B

Nt’ distribution

B

A

Nt’(i) < Nmed?

TRUE

FALSE

SMAS based splitSMAS based split

Nt’(1)Nt’(2)

Nt’(n)

Nt’(3)

Nt’(n-1)

Nt’(1)<..<Nt’(i)<…<Nt’(n)

÷ 2

0

5

10

15

20

25

30

35

40

-12-10

.6666

67-9.

3333

333 -8

-6.66

6666

7-5.

3333

333 -4

Ris

k A

ssess

men

t b

ase

d o

n S

MA

S sp

lit

0

5

10

15

20

25

30

35

-14 -13.3 -12.5 -11.8 -11 -10.3 -9.5 -8.75 -8 -7.25 -6.5 -5.75 -5

2. SMAS based split

69

69

0

5

10

15

20

25

30

35

40

-12.0

-11.3

-10.7

-10.0 -9.3

-8.7

-8.0

-7.3

-6.7

-6.0

-5.3

-4.7

-4.0

0

5

10

15

20

25

30

35

-14 -13.3 -12.5 -11.8 -11 -10.3 -9.5 -8.75 -8 -7.25 -6.5 -5.75 -5

0

5

10

15

20

25

30

35

-14 -13.3 -12.5 -11.8 -11 -10.3 -9.5 -8.75 -8 -7.25 -6.5 -5.75 -5

Typical results

Distant market Local market

log (probability of illness)


0

5

10

15

20

25

-12.0

-10.7 -9.3

-8.0

-6.7

-5.3

-4.0



% o

f cas

es%

of c

ases

% o

f cas

es

Ran

dom

Spl

it%

of c

ases

SMA

S ba

sed

Split

70

70

Microbial growth kinetic models for:Listeria monocytogenes (EA ≅ 94.5 kJ/mol and µref@10°C≅0.058h-1) and

spoilage microorganisms, Pseudomonas (EA ≅ 73.1kJ/mol and shelflifef@0°C≅190h).

For product monitoring and management at the SMAS points:

an enzymatic TTI (VITSAB® Type L10-3) (EA ≅ 158.5 kJ/mol and max response timef@5°C≅150h)

Real data from surveys, for the stages of transportation to the distribution center, the supermarket storage and stocking of the retail fridge cabinets, and the domestic fridge.

After domestic storage, a cooking step was assumed (logreduction ≅3.0±1.5CFU/g) and a dose-response model (Farber et al 1996) was applied for the estimation of the probability of illness.

Distribution Center

Market AssignmentRandom OR

1st SMASPoint

Export market(+72h transport)

Local market(+24h transport)


Transport(24h)

Display Cabinet(6h-18h-30h)*

DomesticFridge

Product with an initial microbialdistribution

Stock DisplayRandom OR

2nd SMASPoint*

T = 5°C

T = 5°C

Final consumption

0

5

10

15

20

25

-3.7-2.3-1.00.41.83.14.55.97.28.610.011.312.714.1

temperature (°C)

% o

f cas

es

Distribution Center

Market AssignmentRandom OR

1st SMASPoint

Export market(+72h transport)

Local market(+24h transport)


Transport(24h)

Display Cabinet(6h-18h-30h)*

DomesticFridge

Product with an initial microbialdistribution

Stock DisplayRandom OR

2nd SMASPoint*

T = 5°CT = 5°C

T = 5°CT = 5°C

Final consumption

0

5

10

15

20

25

-3.7-2.3-1.00.41.83.14.55.97.28.610.011.312.714.1

temperature (°C)

% o

f cas

es

1st Example of SMAS application

71

71

0

5

10

15

20

25

30

35

40

-12.0

-10.7 -9.3

-8.0

-6.7

-5.3

-4.0

Random split

SMAS based split


% o

f cas

es

0

5

10

15

20

25

30

35

40

45

-50 -40 -30 -20 -10 0 10 20 30 40 50


% o

f cas

es

SMAS based split

Random split

Distant market

72

72

One

out

of

a tri

llion

One

out

of

10 b

illio

ns

One

out

of

100m

illio

ns

One

out

of

a m

illio

n

One

out

of

10.0

00

One

out

of

100

% o

f pro

duct

s


05

10152025303540

FIFOFIFO

SMASSMAS%

of p

rodu

cts


73

73

05

1015202530354045

<0 0.6 1.3 1.9 2.5 3.1 3.8 4.4 5.0Time to end of shelf life (days)

% o

f pro

duct

s SMASSMASFIFOFIFOR

EJEC

TED


74

74Safety and quality assessment

Method: Comparison of listeriosis risk and remaining shelf life at the time of consumptionin products managed with FIFO and SMAS system using Monte Carlo simulation

Product: Cooked meat (ham)

Tools: • Microbiological data from literature• Temperature data collected in the chill chain• validated kinetic models of microbial growth • validated kinetic models of TTI response

Decision points Decision points in the chill chain:in the chill chain:

1. Distribution Center2. Stock display (retail level)

2nd Example of SMAS application

75

75

Key point of chill chain

Nt

distribution

A B

Nmed

Export market Local market

Decision

B

A

Nt(i) < Nmed?

TRUE

FALSE

SMAS principles

76

76

Development of a Safety Monitoring and Development of a Safety Monitoring and Assurance System (SMAS) for chilled food Assurance System (SMAS) for chilled food

productsproducts

Evaluation of SMAS effectivenessRisk estimation

Serving: 50 g

Dose responserelationship: Farber et al., 1996

NormalPopulation:

77

77Evaluation of SMAS effectivenessResults

-0.1

0

0.1

0.2

0.3

0.4

0.5

-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1

Log probability of illness

Prob

abili

ty

FIFOSMAS

78

78

Evaluation of SMAS effectivenessProduct Quality (shelf life)

Results

-505

101520253035404550

-15 -10 -5 0 5 10 15 20 25 30

Remaining Shelf Life (Days)

Prod

ucts

%

FIFOSMAS

-2

0

2

4

6

8

10

12

14

-15 -10 -5Remaining Shelf Life (Days)

Prod

ucts

%

FIFO: 14.9%

SMAS: 4.3%

Spoiled products at the time of consumption

79

79EXPERIMENTAL VALIDATION OF SMAS

Product: Cooked Ham (Air)

TTIs: L5-8 & M4-30

Temperature Conditions: ranging from 4 to 12oC

Bacteria measured: Lactic Acid Bacteria, Listeria Monocytogenes

SMAS decsion time: 144hours

SMAS validation

80

80Design of the Field Test

120 packages of ham

Laboratorysimulatedconditions

Distribution center(decision point)

Local market

Export marketconsumers

SMAS validation

81

81Laboratory simulated conditions

96 packages

60 Packages2 TTI tags attached on each package

60 packages without TTIs

1st step

2nd step

T1

T3

T2

Decision point

SMAS based split

random split

3rd step

Microbiological analysis4th step

T1 T2

SMAS validation

82

82Distribution of microbiological growth

0

5

10

15

20

25

30<0

0 to

0,5

0,5

to 1

1 to

1,5

1,5

to 2

2 to

2,5

2,5

to 3

3 to

3,5

>3,5

log (Listeria)

% s

ampl

es

SMASFIFO

0

5

10

15

20

25

30

35

40

45

>3.5 3.5 to 4.5 4.5 to 5.5 5.5 to 6.5 6.5 to 7.5 >7.5

log (Lactic Acid Bacteria)

% s

ampl

es

SMASFIFO

Moving the distribution of both pathogens and spoilage to the left

SMAS validation

83

83

Development of a Safety Monitoring and Development of a Safety Monitoring and Assurance System (SMAS) for chilled food Assurance System (SMAS) for chilled food

productsproducts

Spoilage/risk based Management System

Decision is based on growth predictionof risk posing (or spoilage) organisms

Optimization of quality Minimization of safety risk

Koutsoumanis, Taoukis and Nychas (2005) Int. Journal of Food Microbiol.

84

84

S M A S

Development and application of a TTI based Safety Monitoring and Assurance System for Chilled Meat Products

QLK1-CT-2002-02545

A European Commission Research and Technology Development Project

FIFTH FRAMEWORK PROGRAMMEQuality of life and management of living resources

http://smas.chemeng.ntua.gr

Documents

1 The EU-project SMAS: Safety Monitoring and Assurance